subject:"\[openstack\-dev\] Remove vim modelines\?"

Re: [openstack-dev] Remove vim modelines?

2013-10-29 Thread Joe Gordon

So after over 20 messages in this thread, now what?

From my count we have 3 -, and 12 +1s. Is that enough for consensus? Do we
move modelines lines to the bottom? Put them in every file? Or just remove
them?

https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:noboilerplate,n,z
https://review.openstack.org/#/c/51295/

On Mon, Oct 28, 2013 at 10:54 AM, Joe Gordon joe.gord...@gmail.com wrote:

On Mon, Oct 28, 2013 at 9:35 AM, Alexis Lee alex...@hp.com wrote:

As a new OpenStack contributor, +1 for lower barriers to entry. It's
helpful for OS to inform editor style and I don't believe in the
arguments to remove them.

Why remove them?
* we could shrink our codebase by a little bit.

Why do you want to shrink it? Simplify, sure, but modelines are not
code (not Python anyway). Even so, it's only one line per file.

* Modelines aren't supported by default in debian or ubuntu due to
security reasons: https://wiki.python.org/moin/Vim

As I believe with many Vim users, my .vimrc greatly predates my
involvement with OpenStack and it reflects my personal preferences (like
Termie's). I really wouldn't rely on the defaults surviving initial user
contact.

* Having modelines for vim means if someone wants we should support
modelines for emacs
(
http://www.gnu.org/software/emacs/manual/html_mono/emacs.html#Specifying-File-Variables
)

I don't see a problem supporting Emacs, as the other major UNIX editor.
The slippery slope runs dry very quickly after that, I don't see gedit
users petitioning for inclusion (no offense intended to any gedit
devotees!).

Alternatively Emacs users as self-proclaimed users of the bestest editor
ever could just install: https://github.com/cinsk/emacs-vim-modeline
and brag about how flexible their editor is.

* There are other ways of making sure tabstop is set correctly for
python files, see https://wiki.python.org/moin/Vim. I am a vIm user
myself and have never used modelines.

I am a Vim user and frequently use modelines! I like that they 'stick'
to files so even if a file gets separated from its project, the modeline
remains.

From an OpenStack point of view, that is not an issue. If a file gets
separated then why should we worry about if it has a pep8 friendly tabstop?

If I wanted to work on a project with style contrary to OpenStack's, I'd
need to write project-specific autocommands. It feels right to me that
if a project requires specific style, it should gate it (as OS does) and
provide some informative help as well (the modelines).

We don't just gate without telling people, the first two items in hacking
are read pep8

http://docs.openstack.org/developer/hacking/

OS could provide a vim script to set up project-specific style but this
fails the barrier-to-entry test.

* We have vim modelines in only 828 out of 1213 python files in nova
(68%), so if anyone is using modelines today, then it only works 68% of
the time in nova

Then we should fix that, in accordance with OS style guidelines.
It's also possible those are the high-touch files?

As of writing this email we don't OS style guidelines (hacking) don't
require vim modelines. This thread is trying to resolve this.

* Why have the same config 828 times for one repo alone? This
violates the DRY principle (Don't Repeat Yourself).

Because afaik there's no standard way to inform editors of
project-specific style guidelines. Plus previously mentioned benefit of
stickiness.

PEP8 is hardly project specific.

John Dennis said on Fri, Oct 25, 2013 at 04:12:35PM -0400:
Emacs... requires the per file variables to be the 1st line . So don't
see how vim and emacs specifications will coexist nicely and stay that
way consistently.

What's the problem? Vim has no such limitation, Emacs can have 1st line
+ Vim 2nd.

My personal feeling is you need to have enough awareness to configure
your editor correctly to contribute to a project. It's your
responsibility and our gate tools will hold you to that promise.

That's not very welcoming. If a project want to be particular about
its style, it has a responsibilty to not just gate but also assist in
following that style.

We do try to make it easy to follow the style. Its fully documented here
http://docs.openstack.org/developer/hacking/. Also the modeline we use
is hardly enough to follow the style (80 char limit, spacing between
functions, between classes etc.) And we provide several ways for a user to
run the full set of style tests locally.

* Install hacking locally (pip install hacking), and run 'flake8' - which
has a really nice vim plugin too (https://github.com/nvie/vim-flake8)
* Just run 'tox -epep8'

I should mention that not long ago I spent a good half day reading
OpenStack materials on this is how things must be done, too long imho.

Let's just remove the mode lines, they really don't belong in every
file.

Let's just

Re: [openstack-dev] Remove vim modelines?

2013-10-29 Thread Robert Collins

On 28 October 2013 23:54, Joe Gordon joe.gord...@gmail.com wrote:


 We don't just gate without telling people, the first two items in hacking
 are read pep8

 http://docs.openstack.org/developer/hacking/

The gate is super useful, but it's reactive, not proactive. And good
documentation isn't a replacement for just doing the right thing.


  * Why have the same config 828 times for one repo alone?  This
  violates the DRY principle (Don't Repeat Yourself).

 Because afaik there's no standard way to inform editors of
 project-specific style guidelines. Plus previously mentioned benefit of
 stickiness.


 PEP8 is hardly project specific.

The definition of PEP8 is (moderately) global. The choice to use PEP8
is project specific.


 That's not very welcoming. If a project want to be particular about
 its style, it has a responsibilty to not just gate but also assist in
 following that style.

 We do try to make it easy to follow the style. Its fully documented here
 http://docs.openstack.org/developer/hacking/.  Also the modeline we use is
 hardly enough to follow the style (80 char limit, spacing between functions,
 between classes etc.) And we provide several ways for a user to run the full
 set of style tests locally.

Documentation isn't the same as helping. Yes, it's true that the modeline isn't
sufficient, but it does get some of the key variation project
guidelines tend to have: spaces and tabs.


-Rob


-- 
Robert Collins rbtcoll...@hp.com
Distinguished Technologist
HP Converged Cloud

___
OpenStack-dev mailing list
OpenStack-dev@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] Remove vim modelines?

2013-10-29 Thread Joe Gordon

On Tue, Oct 29, 2013 at 11:17 AM, Robert Collins
robe...@robertcollins.netwrote:

 On 28 October 2013 23:54, Joe Gordon joe.gord...@gmail.com wrote:


  We don't just gate without telling people, the first two items in hacking
  are read pep8
 
  http://docs.openstack.org/developer/hacking/

 The gate is super useful, but it's reactive, not proactive. And good
 documentation isn't a replacement for just doing the right thing.


And what is the right thing?  From what you have said my understand is that
the correct thing is: To keep the barrier to entry for new developers as
low as possible by including modelines in every file so that anyone who
uses vim and uses modelines have the correct tabstop etc. But this isn't
enough to pass gate (
http://www.python.org/dev/peps/pep-0008/#whitespace-in-expressions-and-statements
etc),
so unless someone runs flake8 or 'tox -epep8' locally they will still end
up with the gate failing them. So I don't think the current state of
modelines reduces the barrier to entry in a meaningful way (At the very
least, if we keep them they should be on every file).




   * Why have the same config 828 times for one repo alone?  This
   violates the DRY principle (Don't Repeat Yourself).
 
  Because afaik there's no standard way to inform editors of
  project-specific style guidelines. Plus previously mentioned benefit of
  stickiness.
 
 
  PEP8 is hardly project specific.

 The definition of PEP8 is (moderately) global. The choice to use PEP8
 is project specific.


Yup, That is why I didn't say 'everyone uses PEP8,' when using PEP8 I was
referring to the definition (and not even the python library).




  That's not very welcoming. If a project want to be particular about
  its style, it has a responsibilty to not just gate but also assist in
  following that style.
 
  We do try to make it easy to follow the style. Its fully documented here
  http://docs.openstack.org/developer/hacking/.  Also the modeline we use
 is
  hardly enough to follow the style (80 char limit, spacing between
 functions,
  between classes etc.) And we provide several ways for a user to run the
 full
  set of style tests locally.

 Documentation isn't the same as helping. Yes, it's true that the modeline
 isn't
 sufficient, but it does get some of the key variation project
 guidelines tend to have: spaces and tabs.



 -Rob


 --
 Robert Collins rbtcoll...@hp.com
 Distinguished Technologist
 HP Converged Cloud

 ___
 OpenStack-dev mailing list
 OpenStack-dev@lists.openstack.org
 http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

___
OpenStack-dev mailing list
OpenStack-dev@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] Remove vim modelines?

2013-10-29 Thread Andrew Laski

On 10/29/13 at 10:53am, Joe Gordon wrote:

So after over 20 messages in this thread, now what?

From my count we have 3 -, and 12 +1s. Is that enough for consensus? Do we
move modelines lines to the bottom? Put them in every file? Or just remove
them?

I have no strong feeling either way on keeping or removing them. But if
they're going to be removed from individual files it may be helpful to
add one in the HACKING.rst text. Not as a modeline but as guidance on
configuring vim. And examples could be added for other editors if
that's desired.